Sentiment Analysis of Shared Tweets on Global Warming on Twitter with Data Mining Methods: A Case Study on Turkish...

Area of Science:

• Environmental Science
• Computational Linguistics
• Social Media Analytics

Background:

• Social media data offers insights into public perception of environmental issues.
• Sentiment analysis is crucial for understanding public sensitivity and behavior towards environmental topics.
• Limited research exists on sentiment analysis of Turkish-language environmental discussions.

Purpose of the Study:

• To perform sentiment analysis on Turkish tweets concerning global warming and climate change.
• To evaluate the effectiveness of machine learning algorithms for this task.
• To establish a performance benchmark for future Turkish sentiment analysis studies.

Main Methods:

• Utilized supervised machine learning algorithms, specifically linear and probabilistic classifiers.
• Trained models on a dataset of thirty thousand randomly selected Turkish tweets.
• Analyzed sentiment intensity, categorizing tweets as positive, negative, or neutral.

Main Results:

• Successfully detected sentiment intensity (positive, negative, neutral) in Turkish tweets.
• Compared the performance ratios of different machine learning algorithms.
• Provided quantitative performance metrics for the applied methods.

Conclusions:

• Machine learning methods are effective for sentiment analysis of Turkish environmental tweets.
• The study offers valuable benchmarking data for future research in Turkish natural language processing.
• Highlights the potential of social media data for understanding public opinion on climate change in Turkey.